INDEX

Explanations

specialized formats and codes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Error codes or specific values

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disambiguazione

-0.42

MCA

-0.41

 Viitattu

-0.41

şiv

-0.40

 beginnetje

-0.40

 ویکی‌پدیا

-0.39

 fundido

-0.37

 вікісторінку

-0.37

 voilà

-0.37

ViewFeatures

-0.36

POSITIVE LOGITS

IBOutlet

0.55

XmlAccessType

0.53

LookAnd

0.53

 AssemblyVersion

0.53

Дереккөздер

0.50

extAlignment

0.50

colhead

0.50

XmlAccessorType

0.48

CodeAttribute

0.47

Життєпис

0.44

Activations Density 0.052%

specialized formats and codes

Error codes or specific values

No Comments

No Known Activations